1. Field of the Invention
This invention relates to electrodes that can selectively oxidize or reduce a biochemical in a solution. More particularly, it relates to electrodes that can translate the concentration of a biochemical to an electrical current, or can utilize an electrical current to selectively convert one biochemical to another.
2. Description of the Related Art
Enzyme based biosensors (i.e., electrochemical sensors capable of detecting the concentration of a single biochemical species in a medium containing a diverse mixture of other compounds) are used in an increasing number of clinical, environmental, agricultural and biotechnological applications. Amperometric enzyme electrodes typically require some form of electrical communication between the electrode and the active site of the redox enzyme that is reduced or oxidized by the substrate. However, the electrooxidation of a reduced site or the electroreduction of an oxidized site (the rate, of which is proportional to the concentration of the enzyme substrate) is complicated by the fact that the active site is often located deep inside an insulating protein shell. Thus, redox enzymes such as glucose oxidase do not directly exchange electrons with simple metal electrodes.
Historically, electrical communication between the enzyme and electrode has been achieved through the use of diffusing mediators. The first mediator employed for FAD-enzyme electrodes was the natural substrate of the flavin-linked oxidases, O.sub.2. example, the reaction of glucose oxidase (GO) is EQU GO-FAD+glucose.fwdarw.GO-FADH.sub.2 +gluconolactone (1) EQU GO-FADH.sub.2 +O.sub.2 .fwdarw.FAD+H.sub.2 O.sub.2 ( 2)
and the first commercial amperometric glucose sensors measured either the decrease in O.sub.2 concentration at an oxygen electrode, or the increase in H.sub.2 O.sub.2 concentration at a platinum electrode.
There were several problems associated with such devices: (1) the H.sub.2 O.sub.2 degraded the enzyme. Nature alleviates this problem through the use of a second enzyme, usually catalase, which is present in high concentrations in cells and catalyses the disproportionation of the H.sub.2 O.sub.2 ; (2) the electrode current depended on the concentration of both the enzyme substrates, i.e., both glucose and O.sub.2 ; (3) measurement of the H.sub.2 O.sub.2 concentration required both a highly catalytic electrode (e.g., Pt) and a potential (ca. 0.7 V vs. SCE) substantially positive of the reversible potential for the FAD/FADH.sub.2 couple (E.degree. is approximately equal to -0.4 V vs. SCE). This resulted in large spurious currents due to a number of easily oxidized species in the system to be measured. Because of (2) and (3), the amperometric biosensors were not adequately substance-specific.
The most recent devices have employed small diffusing redox shuttles (Ox/Red) such as ferrocenes, quinones, ruthenium ammines, components of organic metals, and octacyanotungstates. In such electrodes, reaction (1) above is followed by EQU GO-FADH.sub.2 +Ox.fwdarw.GO-FAD+2 Red+2H.sup.+ ( 3)
where the reduced form of the shuttle (Red) is subsequently electrooxidized. Catalase can be added to the system to protect the enzyme from H.sub.2 O.sub.2. The potential at which these electrodes operate is only slightly positive of the formal potential of the shuttle, and a highly active noble metal electrode is no longer required for the reaction. Thus, the spurious currents due to competing species may be reduced. Still, in an oxygen containing medium, there is a competition between the oxidized form of the shuttle (Ox) and oxygen for the reduced form of the enzyme (GO-FADH.sub.2), equations (2) and (3). Thus, the electrode current will be independent of the oxygen concentration only insofar as the shuttle can compete effectively with O.sub.2.
Enzyme electrodes such as those just described generally require that the enzyme and shuttle be confined to the proximity of the electrode surface. The small shuttles commonly employed can, however, readily diffuse through the membranes that are needed to contain the enzyme, but permit the passage of the enzyme's substrate, e.g., glucose. Recently, a polymeric redox "wire" based on the poly(vinyl-pyridine) (PVP) complex of Os(bpy).sub.2 Cl (abbreviated POs.sup.30 ; the bpy of the complex is 2,2'-bipyridine) has been introduced which electrically connects the enzyme to the electrode yet, by virtue of its molecular size, remains confined behind the enzyme-containing membrane. This polycationic redox polymer forms electrostatic complexes with the polyanionic glucose oxidase in a manner mimicking the natural attraction of some redox proteins for enzymes, e.g., cytochrome c for cytochrome c oxidase.
Enzyme electrodes now in use are of several different types. One type of electrode amperometrically measures the oxygen content of gas streams entering and leaving a reactor containing the substrate and its enzyme. If oxygen is involved in the substrate's enzymatic oxidation, its level is depleted and the substrate concentration can be deduced from the decrease in the oxygen content of the gas.
With a second type of enzyme electrode, a natural electroreactive product of the enzyme-catalyzed reaction is amperometrically monitored. For example, the enzymatic reaction of substrates like glucose or lactate with oxygen, catalyzed by some oxidases, produces hydrogen peroxide. Hydrogen peroxide can be electrooxidized and thereby the substrate concentration over a certain range can be translated into a current.
In a third type of enzyme electrode, a non-natural redox couple mediates electron transfer from the substrate-reduced enzyme to the electrode. In this scheme, the enzyme is reduced by its natural substrate at a given rate; the reduced enzyme is in turn, rapidly oxidized by a non-natural oxidizing component of a redox couple that diffuses into the enzyme, is reduced, diffuses out and eventually diffuses to an electrode where it is oxidized. Here again, the oxidation current can be related to the concentration of the substrate. A specific example of such a redox mediator is the ferricinium carboxylate/ferrocene carboxylate couple that diffusionally mediates electron transfer from glucose reduced glucose oxidase to a carbon electrode.
Most natural enzymes are not directly oxidized at electrodes, even if the latter are maintained at strongly oxidizing potentials, without being destroyed. Also they are not reduced at strongly reducing potentials without being decomposed. It has, however, been shown that enzymes can be chemically modified by binding to their proteins redox couples, whereupon, if in the reduced state, they transfer electrons to an electrode. Thus, amperometric glucose sensors have been made with glucose oxidase to which ferricinium/ferrocene functions have been chemically bound. It has also been shown that when redox polycations in solution electrostatically complex polyanionic enzymes, electrons will flow in these complexes from the substrate to the enzyme, and from the enzyme through the redox polymer, to an electrode. Glucose electrodes have also been built with these complexes.
The current produced at a given substrate level can depend on the concentration of the active enzyme molecules. It has been shown that natural reaction products, like hydrogen peroxide, deactivate the enzyme. Enzymes are also continuously denatured. It has been shown that the denaturing of enzymes can be retarded by embedding the enzyme in a rigid three-dimensional polymer structure. It has been suggested that such embedding fixes the protein structure of the enzyme, preventing conformational changes that result in its eventual denaturing. For example, chymotrypsin has been stabilized by embedding it in crosslinked poly(methyl methacrylate).